Military

Hands-on A-RCI shore experience at the dual Sun workstation simulators
 - with inputs from real world training tapes.

by CAPT Claude Barron, USN, and Terence M. Stuckart, STSCM (SS), USN, (Ret.)

The system is unsatisfactory, and the ship is not ready to deploy
- Commodore sends.

What system? Sad to say, it's the Acoustic-Rapid Commercial Off-the-Shelf (COTS) Insertion (A-RCI) Phase I sonar, as implemented recently on a Pacific Fleet submarine. That certainly got the attention of those of us in the operational and acquisition world for whom A-RCI has been one of the most promising sonar developments in decades. What could have gone wrong?
A-RCI Phase I was operational on another ship without serious setbacks. Factory training had revealed some growing pains and minor deficiencies, but had not been deemed "unsatisfactory." Traditional approaches to installing A-RCI and conducting subsequent crew training had been used. What was the problem - and what could be done to fix it?

Under the leadership of Team Submarine, A-RCI has been under development in industry to maintain the acoustic advantage that could be lost to likely adversaries by the deployment of increasingly quiet submarines. The A-RCI initiative is leveraging rapid advances in COTS information technologies to develop a new generation of sonar signal processing hardware based on an open architecture and commercial standards in hardware and software. By teaming fleet operators and industry engineers to define new processing approaches, design new displays, and optimize operator interfaces - and by revolutionizing traditional acquisition processes - we hope to get the best of the best to the fleet as quickly as possible.

In the 1990s, the acoustic advantage U.S. submarines enjoyed over foreign counterparts began to diminish as traditional narrow-band acoustic signatures evolved into more complex signal patterns much more challenging to detect and recognize. The A-RCI sonar system was designed with improved signal processing and display capabilities specifically intended to exploit these more subtle threat signatures, and the A-RCI designers did their job well. However, as in all our earlier systems, the final link in the chain of signal recognition is still the operator, and without operators who can recognize real-world threat signatures, the system is useless. And that's where we found the problem. Incorporating the operator as if he were a component of the system - training him to employ and maintain it - had not been successfully achieved in our initial A-RCI implementations.

A New Approach to Training for A-RCI Phase II

This sobering realization provided the impetus to develop an entirely new training approach in preparation for the first A-RCI Phase II installations, which took place in late April 1999. In essence, there were only six months to "make the operator a component of the system" and to ensure his proficiency in using it. Happily, we managed to carry it off, yielding extraordinary improvements in sonar watchstander performance and leaving high-quality training tools onboard each ship for follow-on use.

In achieving this success, we verified two old Navy watchwords: "It's the crew that makes the ship," and "Go ask the chief." The key was bringing in the Concept of Operations and Operator-Machine Interface (OMI) Support Group (COSG). The COSG is an element of the A-RCI Advanced Program Build (APB) Sonar Development Working Group (SDWG), and it consists of both senior Sonar Chief Petty Officers from the fleet and civilians from academia and industry. The COSG was established primarily to engage fleet operators in the design and development of A-RCI displays and OMIs, but when the requirement to address operational training surfaced, the COSG instantly recognized a new challenge and took charge.

In September 1998, the COSG Chairman, Master Chief Terry Stuckart, convened an impromptu meeting with active duty and retired senior fleet sonarmen to analyze fleet-wide operator proficiency and training issues. In implementing the resulting recommendations, Master Chief Stuckart (of COMSUBDEVRON 12) and Master Chief Mike Clinch and Chief Frank Rule of ONI worked with Commander Submarine Force, U.S. Atlantic and Pacific Fleets (COMSUBLANT and COMSUBPAC) to resolve personnel requirements and justify the need for dedicated in-port and at-sea operator training time. In addition, they consulted with the NAVSEA Program Offices and the Submarine Warfare Directorate (OPNAV N779) to ensure that funding and hardware needs could be met.

The COSG made their biggest impact in their role as teachers.  In addition to their normal day-to-day duties and responsibilities at their parent commands, the chiefs of the COSG formed two-man teams and personally conducted shipboard training on a part-time voluntary basis, providing two weeks training in port and one week at sea for every A-RCI Phase II installation. The curriculum was based on a solid foundation of technical knowledge, and it used one-on-one/over-the-shoulder teaching in an operational, "on-watch" environment.

Diagnosing the Need

In developing the curriculum, it was first necessary to determine the extent of the training deficiency and to develop a benchmark to establish standards and measure improvement. Historically, sonar proficiency was measured operationally by comparing individual Sonar Shacks and assigning relative grades, such as "Average" or "Below Average," to determine readiness to deploy. These relative assessments were often subjective and are not particularly helpful in determining fleet-wide operational proficiency or overall training effectiveness. We needed an assessment tool that would put the sonarman into a repeatable simulation, where his capacity to recognize and assess what was evident in the data could be compared to known values and a ground-truth result.

Thus, Senior Chief Bob Willetts at ONI created an assessment "survey" that consisted of a standard series of validated lofargrams of real-world encounters derived from recordings of existing, legacy sonar systems. One hundred examples were created for the survey, including 20 with contacts of interest, such as foreign submarines and threat torpedoes, as well as 80 traces portraying merchants, fishing vessels, or no contacts at all. These were printed on paper for serial presentation to the sonarmen, much as they might be seen on a sonar display during normal search.

The sonarmen were instructed to work through the paper grams in a process similar to paging through towed array beams to search for sonar contacts. They were instructed to flag those they recognized as containing contacts of interest, to analyze the records for tactical information such as target speed and geometry, and finally, to classify the targets as accurately as possible. The 100 lofargrams were organized into five sets of 20 grams each, where each set represented a different ocean area of the world. Operators were given 2.5 hours to complete the task. The survey was distributed and administered to qualified sonar watchstanders onboard submarines, at training commands, and at submarine squadron and group staffs. Nearly 200 operators, including ACINT Specialists, were tested. The results indicated clear weaknesses, not only with the fleet operators, but also with the sonar instructors themselves.

Coincident with the lofargram survey, COMSUBLANT initiated a separate inquiry at the Naval Undersea Warfare Center (NUWC). Known as the "Lost dB Study," its fundamental purpose was to determine why contact hold times observed in shore-based analysis of tape recordings from at-sea events were much longer than the hold times reported in real time. The study tested fleet sonarmen and ACINT Riders on both legacy and developmental sonar systems to determine if the signal excess actually displayed as "voltage" on the operators' screens was being fully exploited. This Lost dB Study clearly confirmed the results of the proficiency survey and showed that one of the primary causes of hold-time differences was the capability and training of watchstanders.

Since the success of our new sonar systems is still fundamentally dependent on the operators' ability to read lofargrams, it follows that no amount of improvement in display formats can overcome a lack of recognition proficiency. Given these realities, it became very clear that A-RCI Phase II Installation training had to teach skills in both operating the system and analyzing lofargrams. We turned next to developing the training tools and a focused curriculum that would yield major improvements in overall "system" detection performance by teaching both of these aspects. 

Implementing the Training System

We needed to find an appropriate combination of engineering tools and acquisition practices to put key elements of a responsive training system into the right hands at the right time. The critical components included:

• The Towed Array Record/Playback Unit (TARPU), an element-level tape recorder/reproducer installed in the front end of the Towed Array processing string

• Transportable Sun workstations running A-RCI and Advanced Processor Build (APB) tactical software

• Training materials in the form of acoustic tape recordings of real-world contacts formatted for playback on the TARPU

The use of an instrumentation-grade tape player to feed recordings of real-world submarine contacts into the A-RCI system to evaluate algorithms and displays during earlier sea tests had shown what an invaluable training tool a tape recorder could be. When the decision was made to include a TARPU in A-RCI Phase II, the time available for choosing an affordable instrument that could also satisfy technical, environmental, and size requirements was highly compressed. Nonetheless, PMS-4252 and ONI managed to find one. Then, the COTS input signal conditioner intended for Phase II had to be accelerated for the first shipboard installation and interfaced with TARPU in only 16 weeks. They did that too. 

To support the Sonar Division's classroom training while A-RCI Phase II was being installed on the ship, a portable shore-based processor with A-RCI's Phase II tactical software was needed. Two relatively inexpensive dual Sun workstation systems, one for each coast, were purchased by the Program Office for classroom training. These training systems might be called simulators, but in fact, they allowed for authentic presentation of real-world acoustic data for processing by real A-RCI tactical software. Moreover, they provided the flexibility needed to tailor training to the individual needs of each submarine crew and run and re-run sections of the curriculum as needed. 

As basic training materials, ONI supplied real-world, element-level acoustic recordings - a major effort by Senior Chief John Leonatti and Senior Chief Jerry Behnken in searching the ONI data base, reviewing the data, identifying suitable acoustic events, and dubbing the 21 tapes needed for training. Supporting documentation that listed target signature characteristics and significant event times for both target and own-ship maneuvers was prepared by Mr. Dennis Bailey in the form of detailed ground truth reports. These became invaluable guidebooks for the training teams, just as the new tapes assumed a central role in onboard training. They provide the ships an organic asset to train new personnel and to practice gram-reading and system operation with real-world data. The tape series can be periodically refreshed and updated with more up-to-date data, as well as providing a source of near Op-Immediate intelligence. Ships preparing for deployment can obtain recently recorded TARPU tapes from other A-RCI ships returning from patrol and play them back immediately to prepare for their next assignment. 

Developing Training Curriculum and Examinations

A senior member of the Pacific Fleet TRE Team used to say."Michael Jordan did not become a great basketball player sitting in a classroom calculating how to shoot baskets. He became a great basketball player by taking a ball out on the court and shooting baskets."

In other words, practice is the key to proficiency. 

Learning acoustic signal recognition and analysis skills is like learning a language. Classroom basics are necessary, but real proficiency occurs when the student is placed "in-country" and forced to use the language as part of his daily life. The same is true for the skill of obtaining tactical information from sonar displays. Training has to be accomplished using the ship's tactical sonar system, vice a laboratory signal analyzer, and if at all possible, onboard ship under simulated or actual at-sea conditions. Training both at sea and in port by simulating an at-sea watchstanding environment was key to the A-RCI Installation Training philosophy and its success.

The curriculum has three fundamental goals:

• Operational Proficiency - training the operators to operate the new system
  

• Employment Proficiency - teaching the operators, supervisors, and officers how to best employ the system for a given tactical scenario
  

• Signal Recognition - significantly improving each sonarman's proficiency in recognizing contacts of interest and using all available acoustic clues to exploit the target

Although, the classroom training emphasizes practical techniques, it was the unanimous opinion of the COSG that a Theory of Operation module be included to give students an appropriate understanding of rudimentary technical sonar concepts - an area of knowledge that had deteriorated throughout the fleet as badly as signal recognition. Thus, a System Overview and Theory of Operation are presented on the first day. Significant topics include towed array characteristics, adaptive beam-forming, and spectral analysis. The following day's syllabus covers A-RCI "knobology" and familiarizes students with display and system options by demonstrating the A-RCI modes on the Sun workstation, with individual sessions at the controls for each sonarman. Next comes a day devoted to signal recognition and acoustic intelligence, presented by an ONI ACINT Specialist. He demonstrates not only the appearance of signals of interest on the A-RCI displays, but also identifies appropriate options and display enhancements for maximizing recognition. 

After the operators have become familiar with operating A-RCI and recognizing contacts of interest, a day is spent teaching system employment. This module focuses on current tactical doctrine and follows the published A-RCI Operating Guidelines, including recommended system lineups and some of the reasons for deviating from default settings. The last day of the classroom curriculum is spent reviewing salient elements of the week's training, testing, and making presentations to the ships' officers.

Although sharing the "head knowledge" of our best operators with the average sonarman was one of the most significant challenges in creating the A-RCI curriculum, it was even more difficult to make the expertise of the signal processing community accessible to our crews. To formulate an approach, an eclectic mix of sonar engineers and system developers from Navy, industry, and academia presented a three-day seminar for the COSG at the Naval Oceanographic Office in Mississippi. The presentations covered aspects of A-RCI from basic towed array theory to more arcane topics such as spatial vernier, adaptive beamforming, and frequency analysis. Master Chief Gero Shafer and Senior Chief Bill Koshoffer from COMSUBLANT then took on the four-month task of translating these high-level technical lectures into a series of briefs that virtually all sonarmen could digest.

Implementing A-RCI Phase II Installation Training in the Fleet

As shown in the accompanying graphic, A-RCI Phase II will be entering the fleet on an ambitious schedule, and adopting the new training plan has required maximum coordination with the Type Commanders to find both dedicated personnel and at-sea operational time to conduct the training. Wholehearted cooperation from COMSUBLANT and COMSUBPAC assured the quality of the result.

Although production and installation has always been the top priority in implementing A-RCI Phase II, training has finally assumed its rightful importance within the acquisition process, and we can say confidently that the operator is now "a component of the system." This accomplishment resulted from a mutually-supportive team effort in which participants no longer looked at operational training as if it were a relay race in which each member stepped off the track and out of the picture after passing the baton. Instead, A-RCI Phase II Installation Training was integrated into a total team effort from start to finish.

The results have included an improvement of 200% or better in sonar watchstander proficiency, as measured by our standard assessment survey, universal praise from Commanding Officers, and institutionalized procedures for the long term. Additionally, N77 has decided to expand the effort by making both the process and selected personnel a permanent part of the training infrastructure. Today, representatives from the Type Commander staffs and ONI, along with retired ACINT Riders hired for their operational and technical expertise, serve as the core team that will conduct A-RCI Installation Training now and in the future. In addition, preparations are under way to use this team to support the instructors in the schoolhouse and pipeline training programs and to accomplish periodic refresher training on ships at sea.

The bottom line is clear. Today, the A-RCI operator is acknowledged as a vital component of the system and treated as such. And in reaching this point, we have relearned three very important lessons:

• Training should take place in the operator's environment, using the right tools

• Having a teacher - more than just an instructor - is critical

• Performance must be measured against an absolute standard at every stage

Our challenge now is to use our A-RCI training philosophy and lessons-learned to improve both acquisition processes and the fleet's operational proficiency in all of our warfare systems.

>